Adjuvant Chemotherapy for Resected Periampullary Adenocarcinoma

Abstract

To the Editor: Dr Neoptolemos and colleagues reported results from the large, multicenter European Study Group for Pancreatic Cancer (ESPAC)-3 periampullary randomized controlled trial, which demonstrated no significant overall survival benefit from adjuvant chemotherapy in patients with nonpancreatic periampullary adenocarcinoma. We have 2 comments with regard to the trial data. We believe a central pathological review of the primary sites of origin should have been conducted to provide added clarity and further insight into the study. First, the determination of the primary site of origin (epicenter) of the resected periampullary cancers remains a challenge and is best conducted by a gastrointestinal pathologist. In the European Organization for Research and Treatment of Cancer (EORTC) phase 3 study of adjuvant radiotherapy in patients with pancreatic and periampullary carcinomas, central pathology review changed the diagnosis (pancreatic, periampullary, or unknown) in 5% of the cases. We believe the rate might be higher in this study because central pathology review would be required to solve the additional and more challenging problem of determining the specific sites of nonpancreatic periampullary adenocarcinoma (biliary, duodenal, or ampullary). In the ESPAC-3 study, 2 pancreatic pathologists reviewed each pathology report without reviewing histological slides. This has inherent limitations as demonstrated by the inability to properly classify the histological subtypes (pancreaticobiliary, intestinal, or mixed) of ampullary carcinoma in 162 patients (55%). Second, the low frequency of duodenal adenocarcinoma (2%) in this periampullary cohort contrasts with prior periampullary data sets in which duodenal adenocarcinoma was generally more common and represented 25% to 33% of nonpancreatic periampullary adenocarcinoma cases. We believe this reflects the degree of uncertainty with regard to the ascribed primary sites of periampullary adenocarcinoma. Because the site of nonpancreatic periampullary adenocarcinoma was a significant variable in the multivariate modeling, the uncertainty in determining the primary site of the periampullary adenocarcinoma in this study would limit the interpretation of the secondary multiple regression survival analysis. Could the authors also specify if the initiation of adjuvant therapy was restricted by a maximum time postsurgery and if a restaging imaging study was required prior to initiation of therapy?